Mining Prostate Cancer Behavior Using Parsimonious Factors and Shrinkage Methods

Class prediction of gene expression data analysis is a very hot research topic in computer vision nowadays, particularly in the prostate cancer tumor. A problem often encountered in accomplishing this task while using high dimensionality of data, and growing information constituting of prostate cancer tumor combination, is the dimensionality reduction. Where number of genes (variables) is very large compared to the number of samples (observations), makes the application of many prediction techniques very difficult. An appropriate solution of the said problem is the reduction in the number of features which may potentially lead to a more accurate desirable model. A proposed framework to solve this problem is to employ dimension reduction statistical techniques. Successfully used in many areas and applied in statistical-related applications. Various machine learning methods have been used to analyze the high dimensional data for cancer classification. These methods have been shown to have statistical and clinical relevance in variety of cancer diagnosis, prognosis and therapeutic guidance. This paper aims at improving the performance of prostate cancer tumor modeling and mining by employing an integrated procedure that combines the dimension reduction and classification features by data mining technology. The proposed method is effectively applied on the available prostate cancer tumor dataset, and the results are interpreted. The superiority of the proposed method over the well-known methods is also discussed.